The Library of Congress >> Especially for Librarians and Archivists >> Standards

MARC Standards

HOME >> MARC Development >> Discussion Paper List


MARC DISCUSSION PAPER NO. 2018-DP05

DATE: January 17, 2018
REVISED:

NAME: Adding Institution Level Information to Subject Headings in the MARC 21 Bibliographic Format

SOURCE: German National Library

SUMMARY: This paper explores different ways of designating in a MARC record that a subject access field in the 6XX region has been added according to the policy of a specific Cultural Heritage Organization. While the preceding Discussion Paper 2017-DP05 focused on subfield $5 in the 6XX fields of the MARC Bibliographic format, this Discussion Paper introduces a second option by using and extending field 883 (Machine-generated Metadata Provenance).

KEYWORDS: Subject Access Fields (BD); Field 883 (BD); Machine-generated Metadata Provenance (BD); Institution level information (BD); Subfield $5, in 6XX fields (BD); Institution to which field applies (BD)

RELATED: 2012-03; 2017-DP05

STATUS/COMMENTS:
01/17/18 – Made available to the MARC community for discussion.

02/11/18 – Results of MARC Advisory Committee discussion: Acknowledging the use case set out by the DNB, the committee did not reach a consensus on whether this should be accommodated within MARC 21. A straw vote showed no support for the $5 option, but some preference for the 883 option. To code local subject information, the inclusion of a new first indicator value of "2" with label "Not machine generated" would avoid problems of backwards compatibility. DNB will consider submitting a proposal focusing upon the option to broaden the existing scope of field 883.


Discussion Paper No. 2018-DP05: Adding Institution Level Information to Subject Headings

1. BACKGROUND

Discussion Paper 2017-DP05, "Providing Institution Level Information by Defining Subfield $5 in the 6XX Fields of the MARC 21 Bibliographic Format", was discussed in January 2017 at the meeting of the MARC Advisory Committee. The paper was based on the use case that in the same record some fields in the 6XX range can be provided using the same thesaurus (indicator or $2) but with different sets of rules, or with the same set of rules but using different policies, by different institutions. The format solution discussed was the definition of subfield $5 (Institution to which field applies).

Quoting from the section, Results of MARC Advisory Committee discussion: "Some members expressed understanding for the use cases presented by the German National Library. Other members raised concerns, e.g. the risk of introducing inconsistencies in the 6XX block. The usage of subfield $5 within MARC is so persistent that there was widespread reservation regarding this proposed re-use of the subfield. Alternative solutions were discussed: Subfield $1 which may not be available, subfield $2 containing 'local' or containing 'gnd' + some punctuation + the MARC Organization Code, subfield $i which may better suit information from RDA Appendix M, subfield $q with a conflict in field 600, or field 883 (Machine-generated Metadata Provenance) with some changes needed in definition and scope. The German National Library will consider the options, and decide whether to prepare a follow-up paper for the next MAC meeting."

Most of the different new options mentioned above were analyzed, resulting in an estimation that they are not worth further consideration:

The initial approach of adding $5 to the 6XX fields remains to be discussed as option 1.

In addition, option 2 is based on field 883 (Machine-generated Metadata Provenance).

2. DISCUSSION

2.1. Option 1: Subfield $5 in 6XX Fields

Subfield $5 in general is intended to carry item level information ("institution X has this single item with these specific characteristics"), or it is a signal that information in a field is only relevant for a single institution. Routines exist which delete any field from a MARC record as soon as a subfield $5 exists in the field that is different from the MARC Organization Code of the institution which has copied the record. As a result, the definition of subfield $5 throughout the 6XX fields risks the loss of information which may be relevant.

Taking the arguments against option 1, the definition of subfield $5 in fields 6XX, into account, it is still true that in field 655 (Index Term - Genre/Form) subfield $5 (Institution to which field applies) is already defined, available and in use, see http://www.loc.gov/marc/bibliographic/bd655.html and http://experimental.worldcat.org/marcusage/655.html.

Furthermore, as written in 2017-DP05, in the MARC Authority format subfield $5 is defined in some number fields, in some note fields, and throughout the 4XX (See From Tracing Fields), the 5XX (See Also From Tracing Fields), and the 7XX (Heading Linking Entry Fields). Here subfield $5 carries the “MARC code of the institution or organization that has added a tracing, reference, or linking entry field to an authority file record. It is used for institution-specific information that may or may not apply to the universal use of the authority record.” This definition seems to add to the justification of further discussing 6XX $5.

2.2. Option 2: Field 883

Field 883 (Machine-generated Metadata Provenance) was defined in 2012, after Proposal 2012-03 "Data Provenance in the MARC 21 Bibliographic Format" was discussed and approved in the June 2012 MARC Advisory Committee meeting. Field 883 is currently defined in the Bibliographic format as follows:

Field Definition and Scope:
Used to provide information about the provenance of metadata in data fields in the record, with special provision for machine generation. Field 883 contains a link to the field to which it pertains. Intended for use with data fields that have been fully or partially machine-generated, i.e., generated by some named process other than intellectual creation.

The use case described above would show the need of some change in field 883 to be covered:

The name of field 883 may be changed from "Machine-generated Metadata Provenance" to "Metadata Provenance".

The definition of the field may reflect the broader scope of the field and its usage. One suggestion may to change the current field definition and scope described above to:

Used to provide information about the provenance of metadata in data fields in the record, whether it consists of fully or partially machine generated metadata, or intellectually assigned metadata. Field 883 contains a link to the field to which it pertains.

The first indicator may be renamed from "Method of machine assignment" to "Method of assignment", and one code may be added: 2 – "Not machine generated", or "Intellectually assigned", or some similar terminology.

Some of the names and definitions of the single subfields may need changes as well, e.g.:

$a - Generation process (NR)            =>        Creation process (NR)
$d - Generation date (NR)                 =>        Creation date (NR)
$q - Generation agency (NR)             =>        Assigning or generating agency (NR)

Because field 883 has been defined in all five MARC formats, the changes would cover each of them.

It should be noted that as of now, most of the metadata contained in a MARC record is assigned by a cataloger. MARC Organization Codes in field 040 (Cataloging Source) with its subfields $a (Original cataloging agency) and $d (Modifying agency) already provide metadata provenance information on a record level. Some fields, such as 082 and 083, have specified subfields $q to designate an "Assigning agency". With the 883 option offered here, a more general and flexible approach could be taken, including the specific use case described.

It should be mentioned that field 883 and the linked field are connected by a subfield $8 (Field link and sequence number), both of them containing the same numerical value as linking number, and the field link type p for "Metadata provenance". This is described in Appendices A (Control Subfields) of the five MARC formats, e.g. Appendix A in the Bibliographic format. The technique of $8 is not widely in use in the MARC community, and its usage forms some challenges for conversion routines and data handling.

3. EXAMPLES

MARC record for the serial "Scape", based on http://d-nb.info/980689902

Option 1: Subfields $5 in the 6XX fields

LDR XXXXXnas a2200709 c 4500
001 980689902
003 DE-101
008 060731c20069999ne\f||p|\\|||\0||||0eng\c
015 ## $a06,A29,1090$2dnb
016 7# $2DE-101$a980689902
016 7# $2DE-600$a2244990-5
022 ## $a1389-742X
035 ## $a(DE-599)ZDB2244990-5
035 ## $a(OCoLC)440162907
040 ## $aDE-89$bger$cDE-101
041 ## $aeng
082 74 $a710$a300$qDE-101$222sdnb
083 7# $a690$qDE-600$222sdnb
245 00 $aScape$bthe international magazine for landscape architecture and urbanism$cStichting Lijn in Landschap Foundation
264 31 $aWageningen$bLijn$aBasel$bBirkhäuser [[anfangs]]$c2006-
362 0# $a2006 - 2012,1; 14.2014 -
650 #7 $0(DE-588)4034350-9$0http://d-nb.info/gnd/4034350-9$0(DE-101)040343502$aLandschaftsgestaltung$2gnd$5DE-101
650 #7 $0(DE-588)4062110-8$0http://d-nb.info/gnd/4062110-8$0(DE-101)040621103$aUrbanität$2gnd$5DE-101
650 #7 $0(DE-588)7700528-4$0http://d-nb.info/gnd/7700528-4$0(DE-101)1002092892$aLandschaftsarchitektur$2gnd$5DE-91
650 #7 $0(DE-588)4056795-3$0http://d-nb.info/gnd/4056795-3$0(DE-101)040567958$aStädtebau$2gnd$5DE-91
650 #7 $0(DE-588)4056754-0$0http://d-nb.info/gnd/4056754-0$0(DE-101)040567540$aStadtplanung$2gnd$5DE-91
655 #7 $aZeitschrift$2gnd$5DE-101
655 #7 $aZeitschrift$2gnd$5DE-101
655 #7 $aZeitschrift$2gnd$5DE-91
655 #7 $aZeitschrift$2gnd$5DE-91
655 #7 $aZeitschrift$2gnd$5DE-91
776 08 $iOnline-Ausg.$tScape$dBasel : Birkhäuser, 2006-2013$hOnline-Ressource$w(DE-600)2576859-1$w(DE-101)100753544X
776 08 $iNiederl. Ausg.$tBlauwe kamer$dWageningen : Stichting Lijn in Landschap, 1998-$w(DE-600)2174092-6$w(DE-101)026951827
850 ## $aDE-101a$aDE-101b

Option 2: Fields 883 and subfields $8

LDR XXXXXnas a2200709 c 4500
001 980689902
003 DE-101
008 060731c20069999ne\f||p|\\|||\0||||0eng\c
015 ## $a06,A29,1090$2dnb
016 7# $2DE-101$a980689902
016 7# $2DE-600$a2244990-5
022 ## $a1389-742X
035 ## $a(DE-599)ZDB2244990-5
035 ## $a(OCoLC)440162907
040 ## $aDE-89$bger$cDE-101
041 ## $aeng
082 74 $a710$a300$qDE-101$222sdnb
083 7# $a690$qDE-600$222sdnb
245 00 $aScape$bthe international magazine for landscape architecture and urbanism$cStichting Lijn in Landschap Foundation
264 31 $aWageningen$bLijn$aBasel$bBirkhäuser [[anfangs]]$c2006-
362 0# $a2006 - 2012,1; 14.2014 -
650 #7 $83\p$0(DE-588)4034350-9$0http://d-nb.info/gnd/4034350-9$0(DE-101)040343502$aLandschaftsgestaltung$2gnd
650 #7 $84\p$0(DE-588)4062110-8$0http://d-nb.info/gnd/4062110-8$0(DE-101)040621103$aUrbanität$2gnd
650 #7 $85\p$0(DE-588)7700528-4$0http://d-nb.info/gnd/7700528-4$0(DE-101)1002092892$aLandschaftsarchitektur$2gnd
650 #7 $86\p$0(DE-588)4056795-3$0http://d-nb.info/gnd/4056795-3$0(DE-101)040567958$aStädtebau$2gnd
650 #7 $87\p$0(DE-588)4056754-0$0http://d-nb.info/gnd/4056754-0$0(DE-101)040567540$aStadtplanung$2gnd
655 #7 $83\p$aZeitschrift$2gnd
655 #7 $84\p$aZeitschrift$2gnd
655 #7 $85\p$aZeitschrift$2gnd
655 #7 $86\p$aZeitschrift$2gnd
655 #7 $87\p$aZeitschrift$2gnd
776 08 $iOnline-Ausg.$tScape$dBasel : Birkhäuser, 2006-2013$hOnline-Ressource$w(DE-600)2576859-1$w(DE-101)100753544X
776 08 $iNiederl. Ausg.$tBlauwe kamer$dWageningen : Stichting Lijn in Landschap, 1998-$w(DE-600)2174092-6$w(DE-101)026951827
850 ## $aDE-101a$aDE-101b
883 2# $83\p$qDE-101
883 2# $84\p$qDE-101
883 2# $85\p$qDE-91
883 2# $86\p$qDE-91
883 2# $87\p$qDE-91

Note:
“DE-101” is the ISIL for the German National Library, URI
http://ld.zdb-services.de/resource/organisations/DE-101
“DE-91” is the ISIL for the “Technische Universität München, Universitätsbibliothek”, URI
http://ld.zdb-services.de/resource/organisations/DE-91

4. BIBFRAME DISCUSSION

This would need to be analyzed for implications for provenance.

5. QUESTIONS FOR DISCUSSION

5.1. Is the need to provide the information that a specific institution has added a subject heading to a MARC record, according to its policy, sufficiently demonstrated in this paper?

5.2. Which option covers the described need in the best possible way: Option 1: defining subfield $5 throughout the 6XX fields, or Option 2: extending field 883?

5.3. Are there any other potential problems that should be taken into account?


HOME >> MARC Development >> Discussion Paper List

The Library of Congress >> Especially for Librarians and Archivists >> Standards
( 04/12/2018 )
Legal | External Link Disclaimer Contact Us